A Novel Robust Fuzzy Programming Approach for Closed-loop Supply Chain Network Design

Attention to environmental issues in supply chain activities has been taken into consideration due to the increase in public awareness and strict laws related to environmental protection. Initially, only the economic aspects of the supply chain were considered in the network configuration, but with increasing concerns about environmental issues, reverse logistics and closed-loop supply chains were developed. Designing a closed-loop supply chain network plays an important role in reducing costs, improving service levels, and responding to environmental issues. Therefore, the purpose of this study is to design a closed-loop supply chain network taking into account hybrid uncertainties and flexibility in constraints.

In most of the conducted studies about supply chain network design, the types of cognitive and random uncertainties, as well as the flexibility of soft constraints, have not been investigated simultaneously, while the conducted modeling is not able to consider hybrid uncertainty in supply chain parameters in the real world. In this study, to simultaneously consider the hybrid uncertainties and flexibility in constraints, a novel model of robust stochastic, possibilistic, and flexible programming based on Me measurement was developed. In this model, the convex combination of optimistic and pessimistic attitudes of decision-makers was considered in the form of the Me measure, and the modeling was more flexible and realistic.

In the proposed approach, a convex combination of optimistic and pessimistic spectra was considered in the model. The need for subjective and repetitive reviews by decision-makers was eliminated in the model and the level of satisfaction was calculated optimally after solving the problem. On the other hand, due to the robustness of the model, possible deviations, scenario deviations, non-fulfillment of demand and capacity, and deviations of soft constraints were minimized. In the proposed approach based on the Me measure, the problem-solving approach was reduced and there was no need for a two-step solution to find solutions.

A case study was conducted in the supply chain of stone paper production to evaluate the efficiency of the proposed model. The results of sensitivity analysis, robustness analysis, and simulation with the realization model showed that the proposed model was able to provide robust and realistic solutions. The proposal of a realistic and flexible solution for designing problems of the supply chain network by creating a trade-off between the objective function and the risk-taking level of decision-makers and managers through changing the justified space in the Me criterion in the proposed approach was one of the achievements of the present study. As its other achievement, the present study could provide a combination of different viewpoints of decision-makers’ risk-taking through changing the justified space based on different values of the parameter λ in measuring Me and propose flexible and realistic solutions according to the results of numerical simulation in the proposed approach.